Currently, the buffers of RX queue are allocated from the page pool. In
the subsequent patches to support XDP zero copy, the RX buffers will be
allocated from the UMEM. Therefore, extract fec_alloc_rxq_buffers_pp()
from fec_enet_alloc_rxq_buffers() and we will add another helper to
allocate RX buffers from UMEM for the XDP zero copy mode. In addition,
fec_alloc_rxq_buffers_pp() only initializes bdp->bufaddr and does not
initialize other fields of bdp, because these will be initialized in
fec_enet_bd_init().
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-14-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Extract fec_xdp_rxq_info_reg() from fec_enet_create_page_pool() and move
it out of fec_enet_create_page_pool(), so that it can be reused in the
subsequent patches to support XDP zero copy mode.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-13-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Remove the size parameter from fec_enet_create_page_pool(), since
rxq->bd.ring_size already contains this information.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-12-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The tx_buf has three types: FEC_TXBUF_T_SKB, FEC_TXBUF_T_XDP_NDO and
FEC_TXBUF_T_XDP_TX. Currently, the driver uses 'if...else...' statements
to check the type and perform the corresponding processing. This is very
detrimental to future expansion. To support AF_XDP zero-copy mode, two
new types will be added in the future, continuing to use 'if...else...'
would be a very bad coding style. So the 'if...else...' statements in
the current driver are replaced with switch statements.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-11-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The tx_buf pointer will not NULL when its type is FEC_TXBUF_T_XDP_NDO or
FEC_TXBUF_T_XDP_TX. If the type is FEC_TXBUF_T_SKB, dev_kfree_skb_any()
will do NULL pointer check. So it is unnecessary to do NULL pointer check
in fec_enet_bd_init() and fec_enet_tx_queue().
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-10-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Currently, the driver writes the ENET_TDAR register for every XDP frame
to trigger transmit start. Frequent MMIO writes consume more CPU cycles
and may reduce XDP TX performance, so transmit XDP frames in bulk.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-9-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Remove fec_enet_xdp_get_tx_queue() from fec_enet_xdp_tx_xmit() and add
the tx_qid parameter to it. Then, calculate the TX queue ID for XDP_TX
frames in fec_enet_rx_queue_xdp(). This way, the TX queue ID only needs
to be calculated once for XDP_TX frames during each NAPI polling. And
since the number of RX queues and TX queues in FEC is generally equal,
the RX queue ID can be directly used as the TX queue ID. In exceptional
cases, fec_enet_xdp_get_tx_queue() is used to calculate the TX queue ID.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-8-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Currently, the processing of XDP path packets and protocol stack packets
are both mixed in fec_enet_rx_queue(), which makes the logic somewhat
confusing and debugging more difficult. Furthermore, some logic is not
needed by each other. Such as the kernel path does not need to call
xdp_init_buff(), XDP path does not support swap_buffer(), because
fec_enet_bpf() returns "-EOPNOTSUPP" for those platforms which need
swap_buffer()), and so on. This prevents XDP from achieving its maximum
performance. Therefore, XDP path packets processing has been separated
from fec_enet_rx_queue() by adding the fec_enet_rx_queue_xdp() function
to optimize XDP path logic and improve XDP performance.
The XDP performance on the iMX93 platform was compared before and after
applying this patch. Detailed results are as follows and we can see the
performance has been improved.
Env: i.MX93, packet size 64 bytes including FCS, only single core and RX
BD ring are used to receive packets, flow-control is off.
Before the patch is applied:
xdp-bench tx eth0
Summary 396,868 rx/s 0 err,drop/s
Summary 396,024 rx/s 0 err,drop/s
xdp-bench drop eth0
Summary 684,781 rx/s 0 err/s
Summary 675,746 rx/s 0 err/s
xdp-bench pass eth0
Summary 208,552 rx/s 0 err,drop/s
Summary 208,654 rx/s 0 err,drop/s
xdp-bench redirect eth0 eth0
eth0->eth0 311,210 rx/s 0 err,drop/s 311,208 xmit/s
eth0->eth0 310,808 rx/s 0 err,drop/s 310,809 xmit/s
After the patch is applied:
xdp-bench tx eth0
Summary 409,975 rx/s 0 err,drop/s
Summary 411,073 rx/s 0 err,drop/s
xdp-bench drop eth0
Summary 700,681 rx/s 0 err/s
Summary 698,102 rx/s 0 err/s
xdp-bench pass eth0
Summary 211,356 rx/s 0 err,drop/s
Summary 210,629 rx/s 0 err,drop/s
xdp-bench redirect eth0 eth0
eth0->eth0 320,351 rx/s 0 err,drop/s 320,348 xmit/s
eth0->eth0 318,988 rx/s 0 err,drop/s 318,988 xmit/s
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-7-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
This patch has made the following adjustments to fec_enet_rx_queue().
1. The function parameters are modified to maintain the same style as
subsequently added XDP-related interfaces.
2. Some variables are initialized at the time of declaration, and the
order of local variables is updated to follow the reverse xmas tree
style.
3. Replace variable cbd_bufaddr with dma.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-6-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Extract the helper fec_build_skb() from fec_enet_rx_queue(), so that the
code for building a skb is centralized in fec_build_skb(), which makes
the code of fec_enet_rx_queue() more concise and readable.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-5-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The FEC of some platforms supports RX FIFO shift-16, it means the actual
frame data starts at bit 16 of the first word read from RX FIFO aligning
the Ethernet payload on a 32-bit boundary. The MAC writes two additional
bytes in front of each frame received into the RX FIFO. Currently, the
fec_enet_rx_queue() updates the data_start, sub_len and the rx_bytes
statistics by checking whether FEC_QUIRK_HAS_RACC is set. This makes the
code less concise, so rx_shift is added to represent the number of extra
bytes padded in front of the RX frame. Furthermore, when adding separate
RX handling functions for XDP copy mode and zero copy mode in the future,
it will no longer be necessary to check FEC_QUIRK_HAS_RACC to update the
corresponding variables.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-4-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Extract fec_rx_error_check() from fec_enet_rx_queue(), this helper is
used to check RX errors. And it will be used in XDP and XDP zero copy
paths in subsequent patches.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-3-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Currently, the workaround for FEC_QUIRK_ERR007885 has three call sites,
so add the helper fec_txq_trigger_xmit() to make the code more concise
and reusable.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Link: https://patch.msgid.link/20260205085742.2685134-2-wei.fang@nxp.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
When CONFIG_SLAB_FREELIST_RANDOM is enabled and get_random_u32()
is called in an NMI context, lockdep complains because it acquires
a local_lock:
================================
WARNING: inconsistent lock state
6.19.0-rc5-slab-for-next+ #325 Tainted: G N
--------------------------------
inconsistent {INITIAL USE} -> {IN-NMI} usage.
kunit_try_catch/8312 [HC2[2]:SC0[0]:HE0:SE1] takes:
ffff88a02ec49cc0 (batched_entropy_u32.lock){-.-.}-{3:3}, at: get_random_u32+0x7f/0x2e0
{INITIAL USE} state was registered at:
lock_acquire+0xd9/0x2f0
get_random_u32+0x93/0x2e0
__get_random_u32_below+0x17/0x70
cache_random_seq_create+0x121/0x1c0
init_cache_random_seq+0x5d/0x110
do_kmem_cache_create+0x1e0/0xa30
__kmem_cache_create_args+0x4ec/0x830
create_kmalloc_caches+0xe6/0x130
kmem_cache_init+0x1b1/0x660
mm_core_init+0x1d8/0x4b0
start_kernel+0x620/0xcd0
x86_64_start_reservations+0x18/0x30
x86_64_start_kernel+0xf3/0x140
common_startup_64+0x13e/0x148
irq event stamp: 76
hardirqs last enabled at (75): [<ffffffff8298b77a>] exc_nmi+0x11a/0x240
hardirqs last disabled at (76): [<ffffffff8298b991>] sysvec_irq_work+0x11/0x110
softirqs last enabled at (0): [<ffffffff813b2dda>] copy_process+0xc7a/0x2350
softirqs last disabled at (0): [<0000000000000000>] 0x0
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(batched_entropy_u32.lock);
<Interrupt>
lock(batched_entropy_u32.lock);
*** DEADLOCK ***
Fix this by using pseudo-random number generator if !allow_spin.
This means kmalloc_nolock() users won't get truly random numbers,
but there is not much we can do about it.
Note that an NMI handler might interrupt prandom_u32_state() and
change the random state, but that's safe.
Link: https://lore.kernel.org/all/0c33bdee-6de8-4d9f-92ca-4f72c1b6fb9f@suse.cz
Fixes: af92793e52 ("slab: Introduce kmalloc_nolock() and kfree_nolock().")
Cc: stable@vger.kernel.org
Signed-off-by: Harry Yoo <harry.yoo@oracle.com>
Link: https://patch.msgid.link/20260210081900.329447-3-harry.yoo@oracle.com
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Lockdep complains when get_from_any_partial() is called in an NMI
context, because current->mems_allowed_seq is seqcount_spinlock_t and
not NMI-safe:
================================
WARNING: inconsistent lock state
6.19.0-rc5-kfree-rcu+ #315 Tainted: G N
--------------------------------
inconsistent {INITIAL USE} -> {IN-NMI} usage.
kunit_try_catch/9989 [HC1[1]:SC0[0]:HE0:SE1] takes:
ffff889085799820 (&____s->seqcount#3){.-.-}-{0:0}, at: ___slab_alloc+0x58f/0xc00
{INITIAL USE} state was registered at:
lock_acquire+0x185/0x320
kernel_init_freeable+0x391/0x1150
kernel_init+0x1f/0x220
ret_from_fork+0x736/0x8f0
ret_from_fork_asm+0x1a/0x30
irq event stamp: 56
hardirqs last enabled at (55): [<ffffffff850a68d7>] _raw_spin_unlock_irq+0x27/0x70
hardirqs last disabled at (56): [<ffffffff850858ca>] __schedule+0x2a8a/0x6630
softirqs last enabled at (0): [<ffffffff81536711>] copy_process+0x1dc1/0x6a10
softirqs last disabled at (0): [<0000000000000000>] 0x0
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&____s->seqcount#3);
<Interrupt>
lock(&____s->seqcount#3);
*** DEADLOCK ***
According to Documentation/locking/seqlock.rst, seqcount_t is not
NMI-safe and seqcount_latch_t should be used when read path can interrupt
the write-side critical section. In this case, do not access
current->mems_allowed_seq and avoid retry.
Fixes: af92793e52 ("slab: Introduce kmalloc_nolock() and kfree_nolock().")
Cc: stable@vger.kernel.org
Signed-off-by: Harry Yoo <harry.yoo@oracle.com>
Link: https://patch.msgid.link/20260210081900.329447-2-harry.yoo@oracle.com
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
The only thing this driver's init/exit functions do is call
pci_register/unregister_driver, and in the case of the init function,
print an unnecessary message. Replace them with module_pci_driver to
simplify the code.
Signed-off-by: Ethan Nelson-Moore <enelsonmoore@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260205070632.37516-1-enelsonmoore@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Ensure that the pipe BPP range is valid after calculating the minimum
and maximum pipe BPP values separately.
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patch.msgid.link/20260209133817.395823-2-imre.deak@intel.com
The pipe BPP value shouldn't be set outside of the source's / sink's
valid pipe BPP range, ensure this when increasing the minimum pipe BPP
value to 30 due to HDR.
While at it debug print if the HDR mode was requested for a connector by
setting the corresponding HDR connector property. This indicates
if the requested HDR mode could not be enabled, since the selected
pipe BPP is below 30, due to a sink capability or link BW limit.
v2:
- Also handle the case where the sink could support the target 30 BPP
only in DSC mode due to a BW limit, but the sink doesn't support DSC
or 30 BPP as a DSC input BPP. (Chaitanya)
- Debug print the connector's HDR mode in the link config dump, to
indicate if a BPP >= 30 required by HDR couldn't be reached. (Ankit)
- Add Closes: trailer. (Ankit)
- Don't print the 30 BPP-outside of valid BPP range debug message if
the min BPP is already > 30 (and so a target BPP >= 30 required
for HDR is ensured).
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/7052
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/15503
Fixes: ba49a4643c ("drm/i915/dp: Set min_bpp limit to 30 in HDR mode")
Cc: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Cc: <stable@vger.kernel.org> # v6.18+
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com> # v1
Reviewed-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://patch.msgid.link/20260209133817.395823-1-imre.deak@intel.com
The vc4 driver relies on a drm_private_obj, that is initialized by
allocating and initializing a state, and then passing it to
drm_private_obj_init.
Since we're gradually moving away from that pattern to the more
established one relying on a atomic_create_state implementation, let's
migrate this instance to the new pattern.
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-14-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The MSM dpu1 driver relies on a drm_private_obj, that is initialized by
allocating and initializing a state, and then passing it to
drm_private_obj_init.
Since we're gradually moving away from that pattern to the more
established one relying on a atomic_create_state implementation, let's
migrate this instance to the new pattern.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-11-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The MSM mdp5 driver relies on a drm_private_obj, that is initialized by
allocating and initializing a state, and then passing it to
drm_private_obj_init.
Since we're gradually moving away from that pattern to the more
established one relying on a atomic_create_state implementation, let's
migrate this instance to the new pattern.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-10-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The ingenic driver relies on two drm_private_objs, that are initialized
by allocating and initializing a state, and then passing it to
drm_private_obj_init.
Since we're gradually moving away from that pattern to the more
established one relying on a atomic_create_state implementation, let's
migrate this instance to the new pattern.
Acked-by: Paul Cercueil <paul@crapouillou.net>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-9-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The ARM komeda driver relies on a number of drm_private_objs, that are
initialized by allocating and initializing a state, and then passing it
to drm_private_obj_init.
Since we're gradually moving away from that pattern to the more
established one relying on a atomic_create_state implementation, let's
migrate this instance to the new pattern.
Acked-by: Liviu Dudau <liviu.dudau@arm.com>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-8-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The DP tunnel implementation relies on a drm_private_obj, that is
initialized by allocating and initializing a state, and then passing it
to drm_private_obj_init.
Since we're gradually moving away from that pattern to the more
established one relying on a atomic_create_state implementation, let's
migrate this instance to the new pattern.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-6-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The DP MST implementation relies on a drm_private_obj, that is
initialized by allocating and initializing a state, and then passing it
to drm_private_obj_init.
Since we're gradually moving away from that pattern to the more
established one relying on a atomic_create_state implementation, let's
migrate this instance to the new pattern.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-5-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The bridge implementation relies on a drm_private_obj, that is
initialized by allocating and initializing a state, and then passing it
to drm_private_obj_init.
Since we're gradually moving away from that pattern to the more
established one relying on a atomic_create_state implementation, let's
migrate this instance to the new pattern.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-4-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Now that we have an atomic_create_state callback for drm_private_objs,
we can provide a helper for it.
It's somewhat different from the other similar helpers though, because
we definitely expect drm_private_obj to be subclassed. It wouldn't make
sense for a driver to use it as-is.
So we can't provide a straight implementation of the atomic_create_state
callback, but rather we provide the parts that will deal with the
drm_private_obj initialization, and we will leave the allocation and
initialization of the subclass to drivers.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-3-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The drm_private_obj initialization was inconsistent with the rest of the
KMS objects. Indeed, it required to pass a preallocated state in
drm_private_obj_init(), while all the others objects would have a reset
callback that would be called later on to create the state.
However, reset really is meant to reset the hardware and software state.
That it creates an initial state is a side-effect that has been used in
all objects but drm_private_obj. This is made more complex since some
drm_private_obj, the DisplayPort ones in particular, need to be
persistent across and suspend/resume cycle, and such a cycle would call
drm_mode_config_reset().
Thus, we need to add a new callback to allocate a pristine state for a
given private object.
This discussion has also came up during the atomic state readout
discussion, so it might be introduced into the other objects later on.
Until all drivers are converted to that new allocation pattern, we will
only call it if the passed state is NULL. This will be removed
eventually.
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-2-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
Since we're going to move the drm_private_obj state allocation to a
callback, we need to be able to deal with its possible failure.
Make drm_private_obj_init return an error code on failure.
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patch.msgid.link/20260128-drm-private-obj-reset-v4-1-90891fa3d3b0@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The get_sample() function in the hwlat tracer assumes the caller holds
hwlat_data.lock, but this is not actually happening. The result is
unprotected data access to hwlat_data, and in per-cpu mode can result in
false sharing which may show up as false positive latency events.
The specific case of false sharing observed was primarily between
hwlat_data.sample_width and hwlat_data.count. These are separated by
just 8B and are therefore likely to share a cache line. When one thread
modifies count, the cache line is in a modified state so when other
threads read sample_width in the main latency detection loop, they fetch
the modified cache line. On some systems, the fetch itself may be slow
enough to count as a latency event, which could set up a self
reinforcing cycle of latency events as each event increments count which
then causes more latency events, continuing the cycle.
The other result of the unprotected data access is that hwlat_data.count
can end up with duplicate or missed values, which was observed on some
systems in testing.
Convert hwlat_data.count to atomic64_t so it can be safely modified
without locking, and prevent false sharing by pulling sample_width into
a local variable.
One system this was tested on was a dual socket server with 32 CPUs on
each numa node. With settings of 1us threshold, 1000us width, and
2000us window, this change reduced the number of latency events from
500 per second down to approximately 1 event per minute. Some machines
tested did not exhibit measurable latency from the false sharing.
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Link: https://patch.msgid.link/20260210074810.6328-1-clord@mykolab.com
Signed-off-by: Colin Lord <clord@mykolab.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Custom target specifications are unstable, but starting with Rust 1.95.0,
`rustc` requires to explicitly pass `-Zunstable-options` to use them [1]:
error: error loading target specification: custom targets are unstable and require `-Zunstable-options`
|
= help: run `rustc --print target-list` for a list of built-in targets
David (Rust compiler team lead), writes:
"We're destabilising custom targets to allow us to move forward with
build-std without accidentally exposing functionality that we'd like
to revisit prior to committing to. I'll start a thread on Zulip to
discuss with the RfL team how we can come up with an alternative
for them."
Thus pass it.
Cc: David Wood <david@davidtw.co>
Cc: Wesley Wiser <wwiser@gmail.com>
Cc: stable@vger.kernel.org # Needed in 6.12.y and later (Rust is pinned in older LTSs).
Link: https://github.com/rust-lang/rust/pull/151534 [1]
Reviewed-by: Gary Guo <gary@garyguo.net>
Tested-by: Gary Guo <gary@garyguo.net>
Link: https://patch.msgid.link/20260206204535.39431-1-ojeda@kernel.org
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
Merge series "slab: replace cpu (partial) slabs with sheaves".
The percpu sheaves caching layer was introduced as opt-in but the goal
was to eventually move all caches to them. This is the next step,
enabling sheaves for all caches (except the two bootstrap ones) and then
removing the per cpu (partial) slabs and lots of associated code.
Besides the lower locking overhead and much more likely fastpath when
freeing, this removes the rather complicated code related to the cpu
slab lockless fastpaths (using this_cpu_try_cmpxchg128/64) and all its
complications for PREEMPT_RT or kmalloc_nolock().
The lockless slab freelist+counters update operation using
try_cmpxchg128/64 remains and is crucial for freeing remote NUMA objects
and to allow flushing objects from sheaves to slabs mostly without the
node list_lock.
Link: https://lore.kernel.org/all/20260123-sheaves-for-all-v4-0-041323d506f7@suse.cz/
Daniel Golle says:
====================
net: dsa: mxl-gsw1xx: setup polarities and validate chip
Now that common PHY properties make it easy to configure the SerDes RX
and TX polarities, use that for the SGMII/1000Base-X/2500Base-X port of
the MaxLinear GSW1xx switches.
Also, validate hardware in probe() function to make sure the switch is
actually present and MDIO communication works properly.
====================
Link: https://patch.msgid.link/cover.1769916962.git.daniel@makrotopia.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
No check for actually present hardware is being performed in the probe
function of the mxl-gsw1xx switch driver. So even if the switch isn't
present at the configured MDIO bus address the driver wrongly tells the
user that a "GSWIP version 0 mod 0" was found, outputting errors about
PHY capabilities not matching.
Read and validate the chip MANU_ID and PNUM_ID registers and output
information while probing, but return an error and abort probing in case
the hardware is not actually present.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Link: https://patch.msgid.link/3194d3d3bb0b51f08755d392e1fdf7bb6dc49608.1769916962.git.daniel@makrotopia.org
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Configure SerDes (port 4) RX and TX polarities using the newly
introduced generic properties. The polarities are described at the port
level which equals the polarities of the external pins of the chip.
Note that the RX lane is inverted internally and the vendor driver
simply always sets bit GSW1XX_SGMII_PHY_RX0_CFG2_INVERT unconditionally
to end up with the correct (ie. as documented in datasheets) polarity at
the external pins.
In this sense, PHY_POLARITY_NORMAL denotes normal polarity for pins as
documented for the MRQFN 105-pin package (GSW120, GSW125, GSW140, GSW141
and GSW145 all use the same package and have identical pin layouts
except for TP port 2 and 3 being N/C on GSW12x):
pin B18 (TX0_P) positive signal of the differential SGMII data output pair
pin B19 (TX0_M) negative signal of the differential SGMII data output pair
pin B20 (RX0_P) positive signal of the differential SGMII data input pair
pin B21 (RX0_M) negative signal of the differential SGMII data input pair
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Link: https://patch.msgid.link/8bf79b3476e23673fceffbe2bc9d6abc13d132e5.1769916962.git.daniel@makrotopia.org
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reference the common PHY properties so RX and TX SerDes lane polarity
of the SGMII/1000Base-X/2500Base-X port can be configured.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/f556ef8be75e37a2f864b9d905a78962bbe76d18.1769916962.git.daniel@makrotopia.org
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Add compatible strings for PMM8654AU power key and resin support.
These blocks are compatible with PMK8350, so use that as the
fallback.
Signed-off-by: Rakesh Kota <rakesh.kota@oss.qualcomm.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com>
Link: https://patch.msgid.link/20260209-add_pwrkey_and_resin-v2-2-f944d87b9a93@oss.qualcomm.com
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
The plane color pipeline initialization built up multiple colorop blocks
inline, but did not reliably clean up partially constructed pipelines
when an intermediate step failed. This could lead to leaked colorop
objects and fragile error handling as the pipeline grows.
Refactor the pipeline construction to use a common helper for adding
colorop blocks. This centralizes allocation, initialization, and
teardown logic, allowing the caller to reliably unwind all previously
created colorops on failure.
v2:
- Refactor code to avoid repetition (Suraj)
v3:
- s/nvl/xe3plpd (Suraj)
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-10-chaitanya.kumar.borah@intel.com
Switch drm_colorop_pipeline_destroy() to use the driver-provided
destroy callback instead of directly calling drm_colorop_cleanup()
and freeing the object.
This allows drivers that embed struct drm_colorop in driver-specific
objects to perform correct teardown.
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-9-chaitanya.kumar.borah@intel.com
Now that colorops are cleaned from drm_mode_config_cleanup(), remove
drm_colorop_pipeline_destroy() from vkms_destroy().
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-8-chaitanya.kumar.borah@intel.com
Tear down all registered drm_colorop objects during
drm_mode_config_cleanup() by invoking their destroy callbacks.
This ensures proper cleanup of color pipeline objects during DRM device
removal.
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-7-chaitanya.kumar.borah@intel.com
i915 embeds struct drm_colorop inside struct intel_colorop, so the
default drm_colorop_destroy() helper cannot be used. Add an
intel_colorop_destroy() helper that performs common DRM cleanup and
frees intel_colorop object.
This ensures correct teardown of plane color pipeline objects.
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-6-chaitanya.kumar.borah@intel.com
Provide a drm_colorop_funcs instance for vkms color pipeline
objects and hook up the common drm_colorop_destroy() helper as the
destroy callback.
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-5-chaitanya.kumar.borah@intel.com
Provide a drm_colorop_funcs instance for amdgpu_dm color pipeline
objects and hook up the common drm_colorop_destroy() helper as the
destroy callback.
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-4-chaitanya.kumar.borah@intel.com
Some drivers might want to embed struct drm_colorop inside
driver-specific objects, similar to planes or CRTCs. In such
cases, freeing only the drm_colorop is incorrect.
Add a drm_colorop_funcs callback to allow drivers to provide a destroy
hook that cleans up the full enclosing object. Make changes in helper
functions to accept helper functions as argument. Pass NULL for now
to retain current behavior.
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-3-chaitanya.kumar.borah@intel.com
Add a helper that performs common cleanup and frees the
associated object. This can be used by drivers if they do not
require any driver-specific teardown.
v2:
- Add function documentation only before definition (Jani)
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://patch.msgid.link/20260202094202.2871478-2-chaitanya.kumar.borah@intel.com
Jakub Kicinski says:
====================
net: stats, tools, driver tests for HW GRO [part]
Add miscellaneous pieces related to production use of HW-GRO:
- report standard stats from drivers (bnxt included here,
Gal recently posted patches for mlx5 which is great)
- CLI tool for calculating HW GRO savings / effectiveness
====================
Link: https://patch.msgid.link/20260207003509.3927744-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Extend ynltool to compute HW GRO savings metric - how many
packets has HW GRO been able to save the kernel from seeing.
Note that this definition does not actually take into account
whether the segments were or weren't eligible for HW GRO.
If a machine is receiving all-UDP traffic - new metric will show
HW-GRO savings of 0%. Conversely since the super-packet still
counts as a received packet, savings of 100% is not achievable.
Perfect HW-GRO on a machine with 4k MTU and 64kB super-frames
would show ~93.75% savings. With 1.5k MTU we may see up to
~97.8% savings (if my math is right).
Example after 10 sec of iperf on a freshly booted machine
with 1.5k MTU:
$ ynltool qstats show
eth0 rx-packets: 40681280 rx-bytes: 61575208437
rx-alloc-fail: 0 rx-hw-gro-packets: 1225133
rx-hw-gro-wire-packets: 40656633
$ ynltool qstats hw-gro
eth0: 96.9% savings
None of the NICs I have access to can report "missed" HW-GRO
opportunities so computing a true "effectiveness" metric
is not possible. One could also argue that effectiveness metric
is inferior in environments where we control both senders and
receivers, the savings metrics will capture both regressions
in receiver's HW GRO effectiveness but also regressions in senders
sending smaller TSO trains. And we care about both. The main
downside is that it's hard to tell at a glance how well the NIC
is doing because the savings will be dependent on traffic patterns.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/20260207003509.3927744-4-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The logic to open a socket and dump the queues is the same
across sub-commands. Factor it out, we'll need it again.
No functional changes intended.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/20260207003509.3927744-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Count and report HW-GRO stats as seen by the kernel.
The device stats for GRO seem to not reflect the reality,
perhaps they count sessions which did not actually result
in any aggregation. Also they count wire packets, so we
have to count super-frames, anyway.
Reviewed-by: Michael Chan <michael.chan@broadcom.com>
Link: https://patch.msgid.link/20260207003509.3927744-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
